The general goal of this project is to provide insight into the nature and function of critical aspects of human memory, adopting a cognitive neuroscience approach that attempts to link together cognitive and neural processes. The experiments proposed here follow directly from research in our laboratory that has focused broadly on trying to understand how memory operates as a constructive process. Research in both cognitive psychology and neuroscience supports the idea that memory is not a simple recording and reproduction of experience, but instead involves constructive processes that assemble memories from bits and pieces of encoded experiences. These constructive processes often produce accurate memories of past events, but sometimes result in errors and distortions that provide key theoretical insights into the components of the underlying cognitive and neural operations. We are concerned with understanding the nature of both encoding and retrieval processes that contribute to constructive memory. Further, we have recently begun to explore how these constructive processes allow individuals to use memory to imagine or simulate possible future events. This broad concern with understanding constructive memory and its role in guiding future-oriented cognitive and neural processes guides all of our specific hypotheses and experiments. Several experiments will explore how memory is used to construct representations of possible future events, using new methods that we have developed in our recent work. A striking finding is that a number of brain regions traditionally associated with memory - including the hippocampus, a structure often implicated in clinical cases of amnesia - also appear to play a role in constructing simulations of future events. Two experiments will test our ideas concerning the role of the hippocampus in recombining and encoding elements of experience, and recalling simulations of future events. Another experiment will test our hypothesis that hippocampal involvement in associative or relational encoding, which promotes flexible access to information useful for simulating future events, may also be linked to memory distortion. Related experiments will test hypotheses concerning how a larger "core network" of brain regions, and subsystems within that network, support future event simulations, whereas others attempt to gain analytic insight into the functioning of this network by examining repetition-related reductions in activity that we have previously studied in attempting to understand nonconscious priming processes. We will also initiate new studies examining the formation of goal-directed simulations and plans. In a closely related line of work, we propose novel paradigms to illuminate the relation between contextual processing and memory distortion, focusing on regions of the core network involved in simulating future events that have also been linked to contextual processing, and exploring predictions concerning the role of these regions in false recognition of everyday scenes. These experiments should both enhance our understanding of basic memory processes, and could also provide insights relevant to clinical conditions. Memory impairments that occur in various neurological and psychological disorders, and as a result of normal aging, can have devastating impacts on everyday life. Further, problems in thinking about future events have been identified in aging, depression, schizophrenia, and other conditions. Our studies should increase our understanding of how we remember the past and imagine the future, thereby providing the kind of insight that is critical for appreciating how these processes change in clinical conditions and what can be done to treat them effectively.